Mycobacterium tuberculosis infects a third of the worlds' population and TB is the leading cause of morbidity and mortality due a single infectious agent. However, only 5-10% of M. tuberculosis-infected subjects without an underlying immunodeficiency develop disease during their lifetimes. Therefore protective immunity is induced in the majority of subjects. Understanding correlates of protection against M. tuberculosis in humans is needed to better direct efforts in the development of antituberculosis vaccines. The PI suggests that increased susceptibility to M. tuberculosis infection by patients with IFNgamma receptor I deficiency and successful therapeutic use of IFNgamma in refractory mycobacterial infections indicates the importance of IFNgamma in immunity against mycobacteria. However, during the previous funding period the PI found increased levels of IFNgamma in the lungs of patients with TB. Thus, the question is raised as to the effectiveness of IFNgamma and/or involvement of other factors in protective immunity. In the present competitive renewal application, the PI proposes to define pulmonary correlates of protective immunity by comparing several immunological parameters in TB patients, healthy household contacts of TB patients (tuberculin skin test positive), and community control subjects. Bronchoalveolar and blood mononuclear cells obtained from each study subject will be utilized to characterize antigen-specific cytokine induction, killing of M. tuberculosis, CTL activity against M. tuberculosis-infected targets and mediators involved in these effector functions (iNOS, granzyme, perforin, granulysin, FasL) (specific aim 1). According to the PI, this first part of the work should identify correlates of protective immunity as differences between protective immune responses (in skin-test-positive, healthy household contacts) and failed immune responses (patients with TB). Correlates of protection identified in aim #1 should then be used to assess induction of protective immunity and chemokine expression by vaccination of humans with BCG. Two strains of BCG having different efficacy and reactogenicity, and two routes of vaccine administration (oral and intracutaneous) will be compared (specific aim 2). The PI states that this proposal attempts to define new parameters of immunological protection in humans and to rationally assess the impact of BCG strain variation and the route of BCG administration on the induction of protective immunity.